The Scope of Jurisdiction of the CJEU in the Field of Common Foreign and Security Policy for Human Rights Violations and the EU’s Accession to the ECHR: Is Schrödinger’s Cat Alive or Dead

Table of Contents: 1. Introduction. – 2. The consequences of Opinion 2/13: Basket Four and the developments of the Court’s case law in relation to the CFSP. – 2.1. The Problem of Basket Four. – 2.2. Exploring the jurisdiction regarding the CFSP: the developing case law of the Court. – 2.3. Refining the jurisdiction of the Court: Neves 77 and KS and KD. – 3. A proposal for examination: Nic Shuibhne’s systemic understanding of the dimensions of the autonomy. – 4. Through Nic Shuibhne‘s autonomy lens: KS and KD, and Neves 77. – 4.1. The purposive dimension. – 4.2. The substantive dimension. – 4.3. The institutional dimension. – 5. Conclusion.

Abstract: The autonomy of EU law has played a key role in discussions surrounding the European Union's accession to the ECHR. In Opinion 2/13, the CJEU stressed that the Draft Agreement on the EU’s Accession was incompatible with EU law, concluding that it conflicted with the principle of autonomy. Ten years later, delegates from the EU and non-EU Member States of the Council of Europe succeeded in reaching agreement on a revised draft. However, one of the concerns the CJEU formulated in Opinion 2/13, namely the precise scope of its jurisdiction with respect to alleged human rights violations concerning acts, actions, or omissions in matters relating to the CFSP, was not addressed, as the Union decided to resolve this matter internally. On 10 September 2024, the CJEU delivered two judgments on judicial review in CFSP matters, the Neves 77 (C-351/22) case and KS and KD (C-29/22 P, joined cases). It was expected that these decisions would improve the EU's prospects of accession to the ECHR. This paper examines whether the CJEU's findings are consistent with the autonomy of EU law. It relies on Nic Shuibhne’s approach, which highlights that the CJEU has established three dimensions of the principle: the purposive, the substantive, and the institutional principle of the autonomy. Reflecting the cases at hand, the paper assesses whether the Court’s findings correspond to the dimensions introduced by Nic Shuibhne.

Keywords: EU accession to the European Convention on Human Rights – autonomy of EU law – Opinion 2/13 – Common Foreign and Security Policy – Neves 77 – KS and KD.

1.   Introduction

The autonomy of European Union (EU) law is of core importance among the structural principles of EU law. It preserves the essential character of the powers of the Union and its institutions to remain unaltered in line with the Treaties.[1] No wonder Nic Shuibhne describes autonomy as having an existential character among the structural principles of the EU legal order.[2] Nic Shuibhne highlights that through its case law, the Court of Justice of the European Union (hereinafter: the ‘CJEU’ or the ‘Court’) has established three specific dimensions for the principle. First, the purposive dimension of the autonomy of EU law enables the Union to pursue its aims and objectives. Second, autonomy has a substantive dimension that ensures the (constitutional) consistency between EU law and other legal regimes of international law. Third, the institutional dimension of autonomy guarantees the respect of allocation of responsibilities and powers of the institutions of the Union and the Member States.[3] The CJEU articulated these three characteristics of the autonomy of EU law in Opinion 2/13, while it held that the Draft Agreement on the EU Accession to the European Convention of Human Rights[4] (hereinafter: ‘ECHR’) was incompatible with EU primary law.[5]

The findings of the CJEU in Opinion 2/13 are incredibly exciting, if only for that reason, because the Court seized the opportunity to present a keen and detailed examination of autonomy in a negative opinion about a decisive move for European integration. After all, the accession of the EU to the ECHR is a project that has long lingered to its place on the political agenda of European integration, beginning with the first attempts to establish a formal connection between the two in the early 1950s.[6] However, it was well known that the EU’s accession would entail significant legal challenges, ranging from the Union’s lack of competence[7] to adhere to the ECHR (before the codification of such an obligation to the EU Treaties) to the more nuanced legal matters surrounding the accession.[8] These challenges were reflected in Opinion 2/13, in which the Court emphasised that the accession should not circumvent the EU legal order. In particular, the Court warned that the European Court of Human Rights (hereinafter ‘ECtHR’) could, in principle, have acquired the opportunity to review the compatibility of acts, actions, or omissions with fundamental rights in areas where the Court itself lacked jurisdiction. This would have allowed the ECtHR to interpret EU law directly, without prior involvement of the Court, disregarding the special characteristics of EU law.[9] As Advocate General Ćapeta observed in her opinion in the KS and KD cases, ‘every train that may end up in Strasbourg first needs to stop in Luxembourg’.[10]

Still, the Union maintained its aspiration to adhere to the ECHR. As a result, the delegations of the Union and the Council of Europe (hereinafter: ‘CoE’) revised the Draft Agreement. Most of the issues regarding the accession (the so-called ‘baskets’) were resolved, with one exception.[11] Basket Four, concerning human rights protection in the Common Foreign and Security Policy (hereinafter: ‘CFSP’), remained a significant challenge, despite the developing case law of the Court. In Opinion 2/13, the Court explained that it ‘has not yet had the opportunity to define the extent to which its jurisdiction is limited in CFSP as a result of [the provisions of the EU Treaties]’,[12] highlighting that the issue could only be settled once a relevant case came before it. This opportunity arose most recently in Neves 77 (C-351/22)[13] and KS and KD (joined cases C-29/22 P and C-44/22 P),[14] which early commentators suggest effectively addressed the outstanding issues in Basket Four and improved the prospect for EU to the ECHR.[15] Indeed, shortly after the release of these judgments, the Commission decided to initiate the procedure before the Court[16] pursuant to Article 218(11) of the Treaty on the Functioning of the European Union (hereinafter ‘TFEU’).[17]

Reflecting the significant implications of Neves 77 and KS and KD for the autonomy of EU law, this paper examines whether the Court has now thoroughly delineated its jurisdiction in CFSP matters. The core of the problem is the scope of the power of the ECtHR to examine applications from persons and States from all areas of EU law, including the CFSP, and establish possible violations of the ECHR for which the EU may be responsible.[18] However, in line with the EU Treaties, the Court has limited jurisdiction over the CFSP. As the allocation of responsibilities of the EU institutions defined in the Treaties, including the Court, is part of the autonomy of EU law,[19] the matter also concerns this autonomy. However, Opinion 2/13 did not formulate the extension of the Court’s jurisdiction over CFSP. Considering Nic Shuibhne’s approach, it should be more thoroughly examined whether the different dimensions of the autonomy in relation to the CFSP remained insufficiently defined. Given that the scope of this jurisdiction remained an open and pressing matter following Opinion 2/13, the paper further considers whether the Court’s most recent findings align with the concept of the ‘autonomy of EU law’. To borrow from the metaphor of Schrödinger’s cat, whether EU autonomy is ‘dead or alive’ can only be determined by looking inside the CFSP box, that is, by closely examining these two judgments.

To provide an answer to such a question, a two-fold approach is required. First, the paper examines the Court’s case law on the CFSP, with particular attention to Neves 77 and KS and KD. Second, given the complex nature of autonomy, a more systemic approach is needed to determine whether these two decisions effectively solve Basket Four. Accordingly, the paper evaluates the Court’s findings in CFSP matters against the three dimensions of the autonomy of EU law identified by Nic Shuibhne.[20]

In this respect, the paper provides a state-of-play analysis that considers both the deficiencies identified in the Draft Agreement at the time of Opinion 2/13, namely those grouped under Basket Four, and the subsequent development of the Court’s case law. It first examines the relevant jurisprudence of the CJEU, focusing on Neves 77 and KS and KD joined cases, which are widely seen as a breakthrough with regard to Basket Four. It then turns to Nic Shuibhne’s framework, and the three dimensions of autonomy she identifies, in order to capture the complexity of this principle and its (admittedly incomplete and imperfect) character. To this end, the paper analyses the Court’s decisions alongside the opinions of the Advocate Generals, while also considering relevant case law of the ECtHR in relation to these judgments. The paper concludes with reflections on the broader trajectory of the Court’s approach to autonomy.

2.   The consequences of Opinion 2/13: Basket Four and the developments of the Court’s case law in relation to the CFSP

2.1.  The Problem of Basket Four

During the accession process to the ECHR, the special characteristics of the EU legal order, including the Court’s jurisdiction over the CFSP, have been somewhat neglected, which highlights the essence of the matter at hand. In this respect, the significance of Basket Four comes from the past. Under the former pillar structure of the EU, the CFSP enjoyed a special and distinct status.[21] The Lisbon reforms intended to integrate the CFSP more fully into EU law by simplifying the system enshrined in the EU Treaties. However, Article 24(1) of the Treaty on European Union (hereinafter: ‘TEU’)[22] still underlines that the CFSP ‘is subject to specific rules and procedures’, and it continues to limit the Court’s jurisdiction in this area – a situation often described as one of ‘crippled conferral’.[23] Under the EU Treaties, two exceptions qualify this carve-out. First, pursuant to Article 24(1) TEU and Article 40 TEU, the Court has jurisdiction to review the legality of CFSP decisions to ensure that they do not encroach upon the Union’s other competences.[24] Second, under Article 275 TFEU, it has jurisdiction to review the legality of CFSP decisions imposing restrictive measures against natural or legal persons adopted by the Council.[25] As these provisions do not provide a comprehensive delineation of the Court’s jurisdiction in CFSP matters, it is unsurprising that the CJEU underlined in Opinion 2/13 that ‘the Court has not yet had the opportunity to define the extent to which its jurisdiction is limited in CFSP matters’.[26] In addition, the Court explained that, under the Draft Agreement, the ECtHR would be empowered to rule on the compatibility of certain CFSP acts with the ECHR, even though the CJEU itself lacks jurisdiction to review such acts in the light of fundamental rights. Prophetically, the Court stressed that ‘it [was] sufficient to declare that, as EU law now stands, certain acts adopted in the context of the CFSP fall outside the ambit of judicial review by the Court of Justice’.[27] Since the exclusive jurisdiction of the Court to interpret EU law and ensure its observance forms an essential element of the autonomy of EU law,[28] the Draft Agreement was found to disregard these specific characteristics and thus failed to safeguard the autonomy of the EU legal order.[29]

For a long time, the idea of the EU accession to the ECHR was put on ice. However, in 2019, the European Commission made a renewed attempt to advance the process, during which the Union resumed negotiations in the so-called 47+1 Group.[30] The 47+1 Group organised the main outstanding issues into four thematic ‘baskets’. In this context, the Union adopted its negotiating directives in line with Article 218 TFEU. Following intense discussions within the 47+1 Group, the negotiating parties reached a provisional agreement in order to address the concerns raised by the Court in Opinion 2/13.[31] However, they refrained from resolving the question of the Court’s jurisdiction over CFSP acts, the so-called Basket Four, as the Union informed the 47+1 Group that, rather than amending the Draft Agreement further, it intended to resolve this issue internally.[32] The EU also committed to keeping the group informed about developments in this regard.[33]

2.2.  Exploring the jurisdiction regarding the CFSP: the developing case law of the Court

Considering the gap identified in Opinion 2/13, it is unsurprising that the Court sought to further elaborate on the matter of the CFSP. After all, the ‘crippled conferral’ left an uncertain legacy of the former pillar structure.[34] In this regard, Article 24(1) second subparagraph, TEU, Article 40 TEU, and Article 275(2) TFEU provided carve-outs from the jurisdiction of the Court. However, in recent years, the CJEU has sought to impose limitations and to shape the scope of its legal review concerning this policy area. In doing so, the Court has stressed the significance of other horizontal principles and rules of EU law.[35] Regarding the development of the case law on Article 40 TEU (in relation to Article 24(1) TEU), the Court has rarely referred explicitly to the provision, although Advocate Generals in various cases have underlined its importance, including its potential for systematic interpretation.[36] Instead, the Court has consistently argued that the carve-out from its jurisdiction must be interpreted in light of the general scope of the its powers as formulated in Article 19 TEU.

The first step of the Court’s determination of its jurisdiction over the CFSP crystallised in its Rosneft judgment. The Court argued that, as the EU Treaties did not provide any particular means by which an examination under Article 40 TEU should be carried out, Article 19 TEU, which defines the general scope of jurisdiction of the Court, should serve as a basis.[37] As Article 19 TEU provides for the CJEU to give preliminary rulings and to decide on the validity of EU acts adopted by the EU institutions, the Court found in it a significant tool to fill the lacuna in the text of the EU Treaties.[38] It explained that the carve-outs from the Court’s jurisdiction concerning the CFSP should be interpreted narrowly in order to maintain the uniform implementation of EU law and the unity of the EU legal order.[39] Admittedly, relying on the general rules of its jurisdiction instead of the carve-outs in Rosneft, the Court stressed that the structure of the EU Treaties had changed to integrate the CFSP into a general framework of law.[40]

In its subsequent decisions concerning the Agreement with Kazakhstan and the Agreement with Armenia, instead of relying on Article 40 TEU, the CJEU based its reasoning on the centre of gravity test as well as the rules of Article 218 TFEU. Rather than interpreting the provisions of Article 40 TEU, the Court simply examined the scope and content of the contested decisions and the procedural rules of Article 218 TFEU for establishing a position to be adopted in a body set up by an agreement, in order to determine the correct legal basis.[41] It is therefore unsurprising that van Elsuwege underlines that the Court treated Article 40 TEU as a redundant provision.[42] Nevertheless, it seems that the Court intended to streamline its general jurisdiction concerning the interpretation and validity of EU law with its limited jurisdiction regarding the CFSP.

Similarly to the previous argument, the Court also interpreted Article 275 TFEU in light of the content of Article 19 TEU, thereby further exploring its jurisdiction over the CFSP in this respect. In the Rosneft case, the Court argued in abstracto that it would run counter to the objectives of Article 19 TEU and the principle of effective judicial protection to adopt a strict interpretation of its jurisdiction under Article 275 TFEU.[43] In this regard, the Court examined additional matters in relation to the scope of the CFSP. In Elitaliana SpA, the CJEU argued that it is the Commission’s responsibility to implement the EU budget and that the general budgetary rules apply to the expenditure of the European Union Rule of Law Mission in Kosovo (hereinafter ‘Eulex Kosovo’). Thus, public procurement rules related to the award of public contracts, giving rise to such expenditure, fall under the rules of the EU Financial Regulation.[44] Furthermore, in Bank Refah Kargaran, the Court accepted its jurisdiction under Article 275 TFEU for actions for damages relating to CFSP decisions concerning restrictive measures.[45] Taking these circumstances into account, as the Court underlined in Elitaliana SpA v Eulex Kosovo, the carve-outs from the Court’s jurisdiction regarding the CFSP cannot be interpreted so extensively as to exclude such matters.[46] A similar approach was taken with respect to the review of acts of staff management concerning to members of the European Union Police Mission (EUPM) Bosnia and Herzegovina. The Court thus held that these matters remained subject to the Staff Regulations during the period of their secondment to the EUPM and therefore fall within the Court’s jurisdiction.[47]

Considering the aforementioned case law of the Court, there is a clear tendency for the CJEU to take the lead in further eroding the remnants of the pillar structure e Maastricht. As Wessel underlines, since the Court’s role is no longer explicitly excluded from the area of the CFSP, its involvement in this field ‘should not come unexpected’.[48] Accordingly, the tendency in the development of the Court’s case law signifies an effort to overcome the deficiencies relating to its role in the CFSP policy area. Yet, after all these years, the question arises as to how this tendency is evolving today.

2.3.  Refining the jurisdiction of the Court: Neves 77 and KS and KD

To understand the matter at hand, two recent judgments should be mentioned. After all, even the European Commission explained that these decisions represented breakthroughs in addressing the relationship between fundamental rights and the CFSP.[49]

2.3.1. Neves 77

The decision of the Court in Neves 77 is of critical importance for the policy area of the CFSP, as the CJEU had the opportunity to clarify its jurisdiction over restrictive measures, even if the Council did not adopt the CFSP measure, by relying on the general jurisdictional rules applying to the Court. The case had its starting point in the imposition of a fine on Neves 77 Solutions by the Romanian National Tax Administration Agency. The company’s business involves brokering the sale of aviation products, specifically facilitating transactions involving Russian-manufactured radio sets. In the present case, Neves acted as an intermediary for a transaction between SFTE Spetstechnoexport, a Ukrainian company, and an Indian company. This conduct violated Council Decision 2014/512/CFSP of 31 July 2014 concerning restrictive measures given Russia's actions destabilising the situation in Ukraine.[50] Neves contested the decision before the national courts in Romania, arguing that its conduct did not constitute an administrative offence and that the tax authority breached its right to property under the ECHR.[51] The appellate court then referred questions of interpretation of the aforementioned Council Decision to the CJEU.[52]

The referring court asked the Court (1) whether the above-mentioned Council Decision permits a national measure authorising a confiscation of the entire proceeds of a transaction; (2) whether it allows Member States to adopt national measures providing for automatic confiscation; and (3) whether the provisions of the Council Decision apply where the goods constituting military equipment, which were the subject of brokering transactions, were never physically imported into the territory of the Member State.[53] These questions thus concerned the interpretation of the Council Decision itself. However, as the CJEU noted in its ruling, the Council should have adopted a regulation under Article 215 TFEU to give effect to the CFSP Decision, thereby regulating the prohibition on providing brokering services pursuant to Article 2(2)(a) of that Decision.[54] The key question, therefore, was whether the Court had jurisdiction to interpret Article 2(2)(a) of the Council Decision in the absence of the corresponding Regulation,[55] which had not yet been adopted at the time of the facts of the main proceedings.[56]

At first glance, the case may appear to raise rather technical and even marginal questions, and with some exceptions, the literature has tended to focus primarily on its legal arguments.[57] However, the case is indeed significant for the Court’s jurisdiction concerning the CFSP. In this instance, Romania, the Netherlands, and the Council argued that the CJEU lacked jurisdiction under Article 24(1) TEU and Article 275 TFEU, in line with the well-established understanding of the carve-outs related to the CFSP.[58] They maintained that the Court could not extend its jurisdiction to interpret provisions of a general scope relating to the CFSP.[59] The Court, sitting in the formation of a Grand Chamber, disagreed. After reiterating the carve-out system,[60] it explained that the individual nature of restrictive measures against natural and legal persons allows it to exercise jurisdiction on such matters.[61] The Court also underlined the importance of Article 215 TFEU, which gives effect to Council decisions in the CFSP field by linking them to economic or financial sanctions. In this respect, the Court relied heavily on its earlier reasoning in Rosneft.[62] Accordingly, it referred once again to Articles 19 TEU and 267 TFEU to safeguard its general jurisdiction and ensure effective judicial monitoring through the uniform application of EU law.[63] On this basis, the Court concluded that the Council’s failure to implement the CFSP measure in time did not prevent it from having jurisdiction in the case.[64]

2.3.2. KS and KD

As mentioned above, both the Commission[65] and the literature paid even greater attention to the judgment in KS and KD, because the Court addressed the legal accountability of fundamental rights claims. For that, the CJEU presented an ambitious approach by introducing the so-called ‘political question doctrine’, distinguishing between political and strategic choices and the acts of day-to-day management.

In this case, KS and KD – two individuals – had lost family members during the Kosovo conflict, and the murders and disappearances remained unresolved. Eulex Kosovo was tasked with investigating such crimes. KS and KD claimed that the Human Rights Review Panel of Eulex Kosovo (hereinafter the ‘Panel’) had failed to properly investigate the crimes committed against their family members, thereby violating their fundamental rights. During the investigations, the Panel found that Eulex Kosovo had breached the right to life, the prohibition of torture, the right to respect for private and family life, and the right to an effective remedy. The Panel made recommendations to Eulex Kosovo, which it found to have been only partially implemented, and subsequently closed the proceedings. KS and KD then brought an action before the General Court (after an earlier dismissal by the High Court of Justice of England & Wales) seeking the annulment or amendment of the CFSP Joint Action on Eulex Kosovo[66] and the subsequent measures, as these governed the functioning and powers of the Panel. The General Court dismissed the action on the grounds that it lacked jurisdiction, and the two individuals appealed to the Court of Justice.[67]

There were several indications that the case was intended to turn the tide regarding the CFSP. First, similar to Neves 77, the Grand Chamber decided the case. Second, seven Member States intervened in the case[68] together with the Commission, each stressing key aspects of the case in their respective arguments. Third, AG Ćapeta was quite ambitious in her opinion, seeking to put an end to the question concerning the Court’s jurisdiction over the CFSP. The AG observed that there are questions related to the CFSP that have significant political implications. In this regard, the drafters of the EU Treaties intended to leave the Court aside at the level of high politics.[69] However, considering the separation of powers (as part of the rule of law and democracy),[70] such an approach cannot be without limits. Thus, it could not have been the intention of the drafters to allow breaches of fundamental rights concerning the CFSP. This leads the AG to conclude that the carve-out provisions limiting the Court’s jurisdiction cannot apply in cases involving fundamental rights violations.[71] This line of argument is puzzling, to say the least, as it relies solely on the rule of law and democracy as constitutional principles, while failing to mention the principle of conferral – one of the most important legal principles of EU law in this context.[72] Moreover, the interpretation does not take into account the limitations on the CFSP laid down in Article 24(1) TEU, Article 40 TEU, and Article 275 TFEU.

In addition, the AG underlined the potential implications of the decision for the EU’s accession to the ECHR (as she had also done in her opinion in Neves 77). AG Ćapeta emphasised that her interpretation would help determine the limits of the Court’s jurisdiction with regard to the CFSP. Furthermore, it would clarify that whenever a CFSP measure is challenged for a reason that the ECtHR could also examine, the EU courts’ jurisdiction would not be restricted by the carve-outs.[73] This, in turn, would satisfy the requirement of the autonomy of EU law. Finally, AG Ćapeta rejected solutions for CFSP matters, ‘in which possible discrepancies between national courts as to whether CFSP measures are in breach of fundamental rights are decided upon by a court external to the EU legal order’.[74] She therefore proposed that the Court should provide an interpretation of its jurisdiction based on the existing EU rules. With her opinion, the AG only heightened expectations that the judgment would be a turning point concerning the Court's jurisdiction over the CFSP.

The Grand Chamber adopted a less holistic but somewhat more nuanced approach to resolve the question at hand. It rejected the AG’s reasoning based on the rule of law and democracy as constitutional principles of the Union. In this regard, in line with the EU Treaties, the Grand Chamber underlined the significance of the principles of conferral and institutional balance, which are defining principles for the Court’s jurisdiction and set its limitations.[75] Hence, the existence of a fundamental rights claim does not automatically establish the Court’s jurisdiction in CFSP matters. What is even more interesting is that the Grand Chamber relied on ECtHR case law. It stressed that the Court must ensure that the interpretation it gives to Article 47 of the Charter of Fundamental Rights of the European Union (hereinafter, the ‘Charter’)[76] is consistent with the provisions of the ECHR.[77] From a doctrinal point of view, it is understandable as it must be ensured, in line with Article 52(3) of the Charter, that insofar as it contains rights, which correspond to rights guaranteed by the ECHR, the meaning and scope of those rights shall be the same as those laid down by the ECHR.[78] The Court's reasoning is certainly substantial, especially in light of the principle of the autonomy, as it intends to create a bridge here between the two fundamental rights standards.

More significantly, albeit through a not particularly well-structured argument, the Court introduced an additional test to the one already derived from the EU Treaties: where Article 24(1) TEU and Article 275 TFEU do not apply, the Court must assess whether the acts and omissions in questions are directly related to political and strategic choices (as those are more acts of day-to-day management) concerning the CFSP or not. If the acts and omissions in question are not directly related to political and strategic choices, the Court would have jurisdiction to assess their legality.[79] In this respect, a case-by-case analysis should be conducted by the Court itself.[80]

Admittedly, the Court refers to its earlier case law when introducing this new test. However, none of the cited cases had previously articulated such an additional test.[81] Instead, although the text of the judgment does state this explicitly, the Grand Chamber appears to rely on the logic applied in ECtHR case law. More specifically, the judgment refers to the reasoning of the ECtHR that an institutional balance between the executive and the national courts should be maintained regarding the conduct of the State in its international relations. In this respect, the judgment in H.F. and Others v France develops such a test, albeit solely under the aegis of the ECHR.[82] In addition, the French Republic referred to the reasoning in Markovic and Others v Italy, arguing that ‘certain acts of foreign policy fall outside the jurisdiction of the courts’, as those are acts of purely administrative management.[83] Although the aforementioned decision was not cited again in this context, the Court relied on a similar argument in its detailed assessment of the acts at issue the case.[84] Concerning the relationship to the ECtHR, it is also a substantial argument that, however, begs further interpretation.

The above reasoning provides even more intriguing insights into the jurisdiction of the CJEU concerning the CFSP than those emerging from Neves 77, that much is certain.It is therefore understandable that the Commission,[85] as well as the academic commentators soon after the publication of the judgment,[86] have been very optimistic about the positive implications of KS and KD for the EU’s accession to the ECHR. In this sense, a careful examination of the autonomy of EU law is needed.

3.   Using Nic Shuibhne’s systemic understanding of the dimensions of autonomy: a proposal

The need for an in-depth understanding of the nature of the autonomy of EU law has become increasingly prominent in the EU legal academic discourse, especially over the past decade following the CJEU’s Opinion 2/13.[87] As a result, this discourse has identified the autonomy of EU law as a structural principle that distinguishes and safeguards the EU legal order from ‘ordinary international law’, a notion traceable to the Court’s early case law.[88] From a legal-theoretical perspective, autonomy possesses a constitutional nature, guaranteeing the preservation of the special characteristics of EU law as well as the values of the Union[89] – or at least establishing an empirical and legal reality for a distinct and autonomous regime of EU law with particular characteristics.[90]

Moreover, the discourse has highlighted that autonomy also appears as a principle in international legal fora and constitutes an attribute of all international organisations.[91] In this regard, both the Court’s case law and academic writings have identified internal and external aspects of the autonomy of EU law. The latter has received significant attention in recent years, as Opinion 2/13 focused heavily on these external aspects.[92] Some scholars have even sought to capture the evolution of the principle and to provide a more systemic and conceptual understanding of it. For instance, Govaere proposed the so-called balloon dynamic, depicting the EU legal order as a balloon membrane, the structural pressure on which must be avoided to preserve the autonomy of the EU legal order.[93] Similarly, Moreno-Lax conceives autonomy as a door to the external world outside the Union – facilitating communication between different legal orders, rules, and values.[94] In other words, autonomy does not only separate the EU legal order from other regimes but also seeks to provide a legally sustainable connection with them. In essence, all the approaches converge on the idea that autonomy is encoded in the very DNA of the Union.[95]

It is understandable that an abstract construct such as the autonomy of EU law requires intellectual exploration in both the legal and theoretical dimensions. However, the question arises whether the Court’s findings in its case law could provide a standard of evaluation to assess a legal act or provision in relation to the principle of autonomy. In this regard, Nic Shuibhne’s argument provides a solid foundation for such an assessment. Nic Shuibhne echoes earlier discussions on the nature of autonomy but further stresses that the principle is not merely structural; rather, it has become an existential, as it is fundamental to the EU legal order.[96] This conclusion rests on three dimensions of autonomy – purposive, substantive, and institutional – which are derived from the Court’s case law and reveal a developmental cycle of the respective principle:

First, considering the purposive dimension, the distinct legal order ‘enables pursuit of the particular aims and objectives of the Union’.[97] In other words, autonomy shields the legal order from both external and internal interferences, allowing it to uphold its own values and objectives. In this sense, Articles 2 and 3 TEU can be effectively maintained and pursued.[98] This gives the principle an existential dimension, as autonomy contributes to preserving the very foundation of the Union.[99] Second, regarding the substantive dimension, Nic Shuibhne argues – drawing on the case law of the Court – that autonomy provides a constitutional guarantee stemming from an autonomous legal system, which is not prejudiced by an international agreement. In this sense, it ensures an EU-dominant outcome in instances of interaction between EU and other standards.[100] This aspect safeguards consistency between the EU legal order and the international legal order[101] and guarantees the system-specific interpretations and functions of the respective EU measures.[102] Nic Shuibhne stresses that this dimension is articulated specifically in relation to the protection of fundamental rights and is therefore relevant to the EU’s accession to the ECHR.[103] Finally, the institutional dimension of autonomy guarantees the preservation of the powers of the EU institutional framework, particularly the exclusive jurisdiction of the CJEU to interpret EU law and ensure observance of the autonomy of the EU legal system.[104]

In this respect, given the optimistic voices in the public about the decisions providing a breakthrough on the accession of the EU to the ECHR, the question arises whether the Court adequately addressed the different dimensions of the principle of autonomy in Neves 77 and the KS and KD. As such, a more thorough examination is needed.

4.   Through Nic Shuibhne‘s autonomy lens: KS and KD, and Neves 77

Although Nic Shuibhne does not argue in favour of a comprehensive test of the principle of autonomy, the author highlights the three aforementioned distinct dimensions of the principle. These dimensions are therefore suitable to be applied as components of a legal test to determine if an international agreement preserves EU autonomy. Admittedly, the dimensions of the autonomy of EU law identified by Nic Shuibhne are interrelated.[105]

In addition, since the test echoes the findings of the Court in Opinion 2/13, these dimensions could also serve as a basis for assessing the autonomy in later cases. Needless to say, as the Court is aware that the legal community closely monitors its judgments, it intentionally provides messages through its reasoning.[106] In this regard, the relevant case law should be examined to determine whether the Court considers these aspects, either intentionally or unintentionally, and whether it conveys such messages concerning the examination of the aforementioned dimensions.

4.1.  The purposive dimension

Considering the purposive dimension, it should be examined whether the Court stresses in its decisions that the Union intends to pursue its aims and objectives. These objectives are relevant for the interpretation of EU law provisions, as they give context and direction to the Union’s activities without prescribing a detailed general programme.[107] In this respect, the Court emphasises in Opinion 2/13 that the Draft Agreement concerning EU accession to the ECHR should respect the autonomy of the Union in order to allow it to pursue its particular objectives.[108] In this sense, autonomy requires that the interpretation of fundamental rights ‘should be ensured within the framework of the structure and objectives of the EU’.[109] However, the CJEU referred the objectives of the Union only briefly in Rosneft with regard to the Court’s jurisdiction over CFSP matters.[110]

It should be borne in mind, though, that the most recent case law of the Court was more direct in addressing the purposive dimension. In KS and KD, the appellants argued that the General Court had failed to interpret, inter alia, the relevant objectives of the Union enshrined in Article 3(5) TEU.[111] In this respect, the Grand Chamber referred to Article 3(5) TEU, underlining that the Union aims to contribute the protection of human rights in its relations with the wider world.[112] In addition, regarding the CFSP, the CJEU, also invoked Articles 21 and 23 TEU, echoing the enumerated objectives of the policy area. In this sense, the Court reinforced those objectives, including the support of human rights and the principles of international law, as well as the preservation of peace, prevention of conflicts, and strengthening of international security, in accordance with the purposes and principles of the United Nations Charter and the Helsinki Final Act.[113] These provisions concerning the CFSP thus somewhat mirror the general objective of the Union incorporated in Article 3(5) TEU.

When assessing the appellants’ argument, the Court also referred to Article 51 of the Charter and its case law, alongside the rule of law and fundamental rights as values enshrined in Article 2 TEU, to underline the significance of judicial review.[114] Although the Court acknowledged the special nature of the CFSP and the distinctive characteristics of the EU legal order (by referring to its findings in Rosneft), it stressed the objectives of the Union in this context. In other words, the Court expressed the purposive dimension. In the subsequent paragraphs, it assessed this complex issue by examining the respective provisions of the ECHR.[115] This demonstrates that, at least regarding this dimension, the Court intends to use the autonomy of EU law not as a tool to fully separate EU law from the international legal order, but to manage external pressure on EU law.

Considering the broader context of international relations, the crucial function of Article 3(5) TEU – a creation of the Lisbon reforms – is to connect the Union (and EU law) to the international arena and the international legal order.[116] In this sense, as Gaudin outlines, this objective could have further potential for the CFSP in the future.[117] To conclude, the Court seems to keep the purposive dimension in mind in its case law, suggesting that this criterion is to be fulfilled.

4.2.  The substantive dimension

Regarding the substantive dimension, which ensures an ‘EU-dominant’ outcome, clarity is essential for a proper assessment. This includes maintaining consistency between the EU legal order and other legal regimes, as well as prioritising the system-specific attributes of EU measures.[118] In this sense, guaranteeing an EU-dominant outcome also requires understanding the relevant concepts that support the preservation of this dimension of the autonomy of EU law. However, the most recent case law of the Court, in this instance, KS and KD, did not provide much clarity on the substantive dimension of the autonomy.

First, in KS and KD, the central issue concerns the clarification and application of the ‘political question doctrine’.[119] Three sub-questions arise in this context. The first concerns the necessity of such a test. At first glance, the KS and KD may seem to have come out of the blue, seemingly without a reason. However, this is not entirely accurate. As van Elsuwege highlighted, the drafters of the Treaties intended to limit the judicial review concerning the CFSP.[120] Moreover, from the very first cases, the need for some sort of test appeared implicitly.[121] As AG Wathelet explained regarding the Rosneft case, ‘CFSP acts are, in principle, solely intended to translate decisions of a purely political nature connected with implementation of the CFSP’.[122] As a result, the need to separate decisions of a purely political nature in the CFSP arose at least soon after the Lisbon reforms. This issue resurfaced in AG Ćapeta’s opinion in KS and KD cases, echoing the argument about the intention of the drafters of the Treaties.[123] Accordingly, it was logical to invoke the necessity of resolving such a problem.

The second sub-question concerns the reasoning behind the CJEU’s application of the ‘political question doctrine’, as the Court did not provide a comprehensive explanation in KS and KD. It is crucial to understand how the ‘political question doctrine’ contributes to achieving an EU-dominant outcome in cases involving interactions between EU and ECHR standards. By adding an additional layer to examine its jurisdiction beyond Article 24(1) TEU and Article 275 TFEU. In addition, by referring to its case law, the Court merely stated that ‘the jurisdiction of the Court of Justice of the European Union may be based on the fact that the acts and omissions at issue are not directly related to the political or strategic choices made by the institutions, bodies, offices and agencies of the Union in the context of the CFSP, and in particular the CSDP’.[124] In other words, the Court refrained from presenting an argument linking this statement to its previous case law. In this regard, the Court neither explained the rationale for adding another step to its test concerning its jurisdiction over CFSP matters based on the provisions of the EU Treaties mentioned above, nor clarified the origin of this second step in its case law.[125] In this respect, this paper lists the following comments.

First, as it appears to be a paradigm-shifting decision concerning CFSP, it requires a more comprehensive explanation, particularly because the Court introduced a separate step to assess its jurisdiction. The Court does refer to its previous cases, but they do not provide any further explanation that would underpin this step.[126]

Second, as noted, it is not the first time that the matter of strategic and political choices in relation to the CFSP policy area has arisen before the CJEU. In H, the issue of ‘administrative acts’ arose as both the appellant and the Commission argued in favour of distinguishing administrative acts, albeit under the domain of carve-out provisions of the EU Treaties.[127] More specifically, the case concerned the exclusion of decisions of an administrative nature from the CFSP area pursuant to Article 24(1) TEU and Article 275(1) TFEU. In this regard, AG Wahl argued that CFSP provisions ‘are necessarily concerned with the dos and don’ts of those administrations. By its very nature, the CFSP appears to be an operational policy: one by means of which the Union pursues its (broadly defined) objectives through a set of (broadly defined) actions, mainly of an executive and political nature’.[128] In this regard, AG Wahl disagreed with the Commission’s arguments to distinguish between ‘acts of sovereign foreign policy’ and ‘acts of implementation’.[129] The Court hinted at this distinction in its judgment concerning the acts of staff management, but, again, only to decide whether the acts fall under the CFSP.[130] Thus, case H does not provide a strong argumentative basis for the Court. This contradiction in the case law (whether apparent or real) should have been addressed in KS and KD.

The third sub-question (taking the argument in the previous paragraph into account) concerns legal certainty. It should be stressed that the literature has identified this problem regarding the possible application of the ‘political question doctrine’. As Lonardo underlined, the moment the ‘political question doctrine’ is introduced into EU law, issues of legal certainty inevitably arise. In this sense, the author predicted a more active role for the CJEU in this field, which, as this question remains unresolved, is yet to be seen.[131] Again, the Grand Chamber hints in KS and KD that the principle of conferral and the principle of institutional balance could hold the key to understanding the doctrine in depth.[132] Yet, considering the grave fundamental rights question at stake (including breaches of the right to life,[133] prohibition of torture,[134] and the right to an effective remedy[135]), the Grand Chamber refrained from doing so. Regardless, such an explanation would be necessary to demonstrate the functioning of the autonomy of EU law in this case.

In this respect, the set of criteria regarding political and strategic choices should be highlighted, as it underscores the precise nature of the various acts concerning the CFSP. To understand the EU-dominant outcome, it is essential to clarify the nature of the different concepts of political and strategic choices, as well as the acts of day-to-day management. As Tzevelekos underlines, a broad interpretation of the notion could limit the jurisdiction of the Court,[136] which would significantly affect the substantive dimension of the autonomy of EU law.

As the Grand Chamber declined AG Ćapeta’s proposal to define the Court’s jurisdiction over cases concerning fundamental rights in general, regardless of the nature of the decisions, it found that the line between political and strategic choices and acts of day-to-day management is crucial. Narrowing the notion could provide a more precise understanding of the Court’s scrutiny.[137] Unfortunately, the Court stressed that whether an act is a political or strategic choice, or belongs to day-to-day management, must be decided on a case-by-case basis, despite the Council’s request for the Grand Chamber to provide a set of criteria.[138] Instead, the Grand Chamber offered no standards to guide assessments of fundamental rights-related challenges. In this respect, it should also be noted that AG Wahl criticized the Commission regarding the distinction between the two groups. As the AG underlined in H, ‘the Commission was at pains to provide some criteria or principles which could be used to distinguish between those two categories of acts’, a position the AG found unconvincing.[139] However, the Court did not address this aspect of the case. Once again, KS and KD was not the first instance in which this problem arose.[140] Unfortunately, the Grand Chamber judgment in KS and KD did not resolve this question, leaving a gap in the assessment of the substantive dimension of autonomy. Understanding that the law must truly reflect the challenges posed by real-life situations, which, at first glance, may justify a case-by-case approach, it may also be necessary to set at least a minimum criterion to ensure predictability in sensitive matters, such as foreign and security policy considerations. Both cases thus represent missed opportunities to properly examine this dimension.

Third, the reference to ECtHR case law also adds some excitement to the matter. As outlined, it is not unfamiliar to the CJEU to separate the roles of the judiciary and the executive in foreign policy decisions, implicitly recognizing the ‘political question doctrine’ in the context of the CFSP.[141] Puzzlingly, the Grand Chamber did not fully highlight this case law when forming the additional test concerning political and strategic choices. Instead, in KS and KD, the Grand Chamber relied on H.F. and Others as a supportive argument regarding institutional balance.[142] This reasoning is not unique, as the ECtHR considered similar matters in various cases.

Two observations should be made here. First, the ECtHR case law applies the doctrine to different rights enshrined in the ECHR. In H.F. and Others, the ECtHR considered the prohibition of expulsion of own nationals under Article 3(2) of Protocol No. 4 to the ECHR.[143] In Tamazount and Others, the ECtHR applied the doctrine in relation to the right to a fair trial under Article 6(1) ECHR, which is more closely aligned with KS and KD.[144] Strikingly, the Grand Chamber did not mention this latter decision. The findings in Tamazount and Others align with the developing case law of the CJEU, underscoring that the right to a fair trial is not absolute. In this context, the Court does refer to Markovic and Others,[145] which illustrates the constitutional limitation of the right to a court under the aegis of the ECHR.[146] The ‘political question doctrine’ thus appears to be a useful tool for different human rights, although the ECtHR case law provides only two specific instances. Considering the particularities of the EU legal order, as highlighted in Opinion 2/13, for instance, further examination of the application of additional fundamental rights would be highly recommended.

Second, it should be noted that the Grand Chamber relied on a doctrine frequently applied in state affairs. The ‘political question doctrine’ exists in both civil law and Anglo-Saxon legal systems.[147] At first glance, applying an approach developed for state affairs might seem far-fetched for the Court. However, in the context of EU external relations law, such an approach is not entirely surprising. The Court has previously compared the common commercial policy to the commercial policies of the Member States.[148] In other words, given the development of EU external relations law, it is not unprecedented for the Court to treat this policy area as one akin to a state’s domain, albeit inconsistently and unpredictably.[149] Again, further explanation from the Court is required to properly assess these factors in relation to the substantive dimension of the autonomy of EU law.

4.3.  The institutional dimension

Although Neves 77 plays a minor role compared to KS and KD regarding the purposive and the substantive dimensions of the autonomy of EU law, both cases are relevant for the institutional dimension.

With regard to CJEU jurisdiction, Neves 77 continued to develop the doctrine outlined in Rosneft and Bank Refah Kagaran. First, the Grand Chamber reiterated in Neves 77 that the derogation from the Court’s jurisdiction should be narrowly interpreted in line with the general jurisdictional powers pursuant to Article 19 TEU, in order to ensure that, in the interpretation and application of the Treaties, the law is observed.[150] In this regard, the Grand Chamber also stressed the need to ensure the consistency of the system of judicial protection under EU law, referring to Article 2 TEU, and the essence of the rule of law. In this sense, the Grand Chamber echoes the reference to the procedures of the Court already clarified in its case law concerning the CFSP, namely preliminary references,[151] its jurisdiction to rule on harm allegedly caused by restrictive measures,[152] and annulment proceedings under Article 263 TFEU.[153] In addition, referring to the further interpretation of Article 215 TFEU, the Court continued its work on filling the gaps. After all, it presupposes that the Court would use the same approach in relation to its other respective proceedings. As such, the fact that the Council fails to adopt the necessary decision to implement a CFSP decision does not hinder the application of the Court’s jurisdiction.[154] As Sarmiento and Sánchez explain, ‘[t]his development closes the circle that remained unfinished after Rosneft’.[155]

However, neither Neves 77 nor KS and KD help to clarify the exact content of Article 40 TEU.[156] As underlined, the Court rarely relies on Article 40 TEU; therefore, the precise content remains uncertain. One can only presume that the provision was intended by the authors of the founding Treaties to exclude certain actions from the Court’s jurisdiction. From the point of view of the institutional dimension of the autonomy of EU law, this statement lacks reasoning after Neves 77 and KS and KD. Again, the former does not address the problem and only briefly refers to Article 40 TEU.[157] As Breitler underlines, Neves 77 was the first case after the Lisbon reforms for which Article 40 TEU should have been decisive.[158] In this sense, the Court referred to Article 40 TEU to stress the CJEU’s ability to monitor compliance with the implementation of the CFSP. In this sense, it does not affect the application of the procedures and the extent of the powers of institutions laid down for the exercise of the Union competences.[159] However, as Breitler rightfully points out, instead of examining Article 2(2)(a) of Council Decision 2014/512/CFSP pursuant to Article 40 TEU to provide guidance on the interpretation of the provision, the Court referred to the examination of the Regulation adopted by the Council based on the Art 215(1) TFEU.[160] As a result, the Court concluded that ‘the jurisdiction of the Court to ensure the judicial protection of third parties cannot be impaired by the Council’s failure to take all necessary measures based on Art 215(1) TFEU’.[161] Again, the Court only mentioned Article 40 TEU but refrained from analyzing its content. From the institutional dimension, Neves 77 is a missed opportunity.

At the same time, KS and KD concerns the ‘political question doctrine’ within the CFSP, not between the CFSP and other policy areas.[162] With regard to KS and KD, it is even more confusing, as the Grand Chamber hinted at the use of historical interpretation, which could have provided the Court with a chance to analyse the intention of the authors of the provision, referred to earlier in this paper.[163] However, the Court did not refer to such legislative history, unlike in its previous case law, where the CJEU explicitly cited its sources.[164] As Article 40 TEU is still considered a provision of strategic importance, it would require further explanation by the CJEU.

Considering the aforementioned test, the institutional dimension should also be highlighted. As underlined in the context of the substantive dimension of the autonomy of EU law, the Court relied on the case law of the ECtHR to identify the appropriate doctrine. Still, nothing guarantees that the content of the doctrine would be identical under the aegis of the two courts. It carries the risk that the ECtHR could have jurisdiction over situations on which the CJEU lacks jurisdiction, simply because it applies different standards for separating political and strategic choices. Again, the Court emphasised that such issues should be assessed on a case-by-case basis,[165] but it cannot guarantee that its exclusive jurisdiction will be maintained. In this regard, the substantive dimension of autonomy correlates with its institutional dimension.

Finally, considering that the Court echoes Lonardo’s approach of refraining from providing a clear-cut criterion to distinguish political and strategic choices from acts of day-to-day management, there is a more or less unexpected consequence. The Court and Lonardo both stress the weight of the principle of conferral, and especially the principle of institutional balance, as both legal principles apply to the relations of the Court.[166] However, it should be stressed that these principles apply to the entire institutional framework of the Union enshrined in Article 13(1) TEU.[167] At the end of the day, the missions of the Union in the CFSP and CSDP policy areas need to apply human rights standards in their work. In this regard, not every situation is clear-cut, making it difficult to distinguish between strategic and political choices and acts of day-to-day management.[168] It is simply not enough to delimit the jurisdiction of EU courts, as the future practice of the doctrine could and will affect everyday work. This could result in hindrances for these missions. Considering the current international geopolitical context and international security, it could produce unexpected outcomes that go beyond the accountability of judges.[169] Again, the question remains whether actors other than the CJEU within the institutional framework could preserve their powers.

As was underlined in relation to Article 40 TEU, which applies to the institutional dimension of the autonomy of EU law, neither of the two cases helped to clarify the normative content of the CJEU’s jurisdiction over the CFSP. Instead, the Court, continuing its habit of appearing in CFSP-related cases, shrouded itself in silence. It seems worrying considering the arguments stressed in Opinion 2/13. The Court expressed that its jurisdiction was yet to be explored, requiring certainty concerning the policy area. However, the ongoing obscurity, which has been echoed in Neves 77 and KS and KD, might risk a permanent entrenching of uncertainty that is least desirable in a changing geopolitical context. Thus, the institutional dimension still requires further clarification.

5.   Conclusion

Considering the CJEU’s evolving case law regarding its jurisdiction over acts, actions, and omissions in the context of the CFSP, the Court has sought to explore the extent of its ability to review the legality of such acts, an issue closely linked to the autonomy of EU law as explained in Opinion 2/13. Both Neves 77 and KS and KD stressed this issue, generating high expectations that they would finally resolve the long-standing problem of Basket Four.

It should be stressed that the aforementioned cases differ significantly. Neves 77 provided a clear-cut approach, aiming to complete the Court’s previous reasonings in Rosneft and Bank Refah Kargaran by balancing the general scope of the CJEU’s jurisdiction under Article 19 TEU with the provisions that carve out exceptions concerning the CFSP.

By contrast, KS and KD introduced an explicit new approach – though one rooted in earlier case law. The case added additional complexities, including, inter alia, the distinction between political and strategic choices and acts of day-to-day management, as well as the practical implications of testing the rest of the institutional framework, as reflected in Article 13(1) first subparagraph TEU. The judgment itself leaves some dissatisfaction for two reasons. First, it merely unveiled the test, refraining from giving in-depth guidance for its consistent application. Second, the text presents a complex jigsaw puzzle: the crucial elements of the Grand Chamber’s reasoning are scattered across different parts of the judgment, requiring careful reconstruction by the reader.

Both cases provided new insights, highlighting further potential developments for the CFSP, although they arguably neglected aspects such as the content of the notion ‘common’ and the role of the Member States. Similar gaps can be detected regarding Article 40 TEU, which the Court effectively overlooked.

Applying the test proposed in this paper, the findings are particularly intriguing. The analysis considered the different dimensions of the autonomy of EU law as proposed by Nic Shuibhne. However, only the purposive dimension received satisfactory treatment in the two cases, aligning with the existential principle of EU law. The substantive and the institutional dimensions, by contrast, remain largely unaddressed. Considering the requirements formulated in Opinion 2/13, future case law – or a potential opinion on the EU’s accession to the ECHR – should engage more thoroughly with both the substantive and the institutional dimensions.

In the current state of affairs, the situation resembles Schrödinger’s cat. In light of the Court’s findings in Neves 77, and particularly in KS and KD, the autonomy of EU law appears to be simultaneously alive and dead in the sealed box of the CFSP. It is the Court’s duty to open the box and to clarify the status of its jurisdiction over the CFSP. In other words, it can only be understood through further observation and measurement of the autonomy of EU law in relation to the CFSP.